The Dimensional Side to Personality

We have been looking at Take Two (from June 21) in the DSM-5’s proposed update to personality disorders. In Take One (from Feb 2010), the DSM-5 attempted to combine categories (as in “which one?) with dimensions (as in “how much?”) into a hybrid system, using interchangeable parts that could best be described as modular. The concept was far-reaching. The catch was that it appeared to be very unwieldy in practice.

Yesterday, we looked at how the DSM-5’s Take Two tackled the categorical side of personality, using a dimensional twist. Today, we look at the DSM-5’s Take Two of the dimensional model, built on categorical pieces, but a different set of categorical pieces from Take One.

Take One, in essence, was based on a sort of reverse Five-Factor Model (such as “antagonism” in place of “agreeableness”) measuring 37 different trait facets on a four-point scale. Take Two is far more modest in scale, but may be much more useful.

Consider, for instance, the following four individuals:

Jane, 28, displays classic borderline symptoms - unstable self-image, trouble maintaining relationships, and so on. But she is working as a manager in a city government. Does the borderline label really apply?

Bill, in his 20s, flies off the handle, uses drugs, and has run-ins with the law, but does not meet the criteria for antisocial. But clearly, his behavior warrants clinical attention.

Sally, 14, is acting out like someone with borderline, but is her behavior more attributable to being a teen?

Joey, in his 50s, deals with major depression and diabetes, and has major issues getting along with his caregivers. Is there a diagnosis that doctors can apply to patients they don’t like?

Enter the Levels of Personality Functioning Scale. This employs the self and interpersonal functions from Criterion A used in all six categorical diagnoses and rates them in terms of severity from 0 to 4. Thus, in 0 (well-adjusted), we see these kind of qualities: “ongoing awareness of a unique self”, “sets and aspires to reasonable goals”, “capable of understanding others”, “maintains multiple, satisfying, and enduring relationships”, and so on.

Meanwhile, way over on 4, we see “boundaries with others are confused and lacking”, “poor differentiation of thoughts from actions” - well, you get the picture.

Then we’re asked to consider the six (mostly interchangeable) trait domains that form the basis of the six personality disorders: Negative Affectivity, Disinhibition, Antagonism, Psychoticism, and Compulsivity. Only this time we are viewing these domains as stand-alone entities rather than in the context of a full-blown personality disorder.

Let’s return to our individuals, who represent abbreviated versions of the examples served up by the DSM-5.

Jane, according to the DSM-5, would rate a diagnosis of borderline, “but her level of personality functioning might be rated as less impaired than that of the more typical borderline patient, with enhanced prospects for successful treatment.” In other words, Jane is a good prognosis patient.

Bill, under the old DSM, would probably sail under the diagnostic radar or fall through the cracks. Perhaps he would be diagnosed as Personality Disorder NOS (not otherwise specified), which tells us nothing. Under the DSM-5, we are told, Bill would be coded as “Personality Disorder Trait Specified,” emphasis on trait specified, such as hostility and impulsivity. These traits would then “serve as specific foci of clinical attention.”

Sally may show signs of emerging borderline, but the DSM-5 indicates the wise course is to hold off on this diagnosis, and instead note her as having a low level of personality functioning, with reference to specific traits such as emotional lability. These features can then be closely tracked as Sally matures.

Joey may not have a personality disorder, but he could be written up for “antagonism,” and impairment in interpersonal personality function. The DSM-5 doesn’t say this, but the best way of treating Joey’s depression and diabetes is to treat the personality issues that sabotage his being a successful patient.

***

Thus, in terms of functional impairment and various traits, in all four cases we are seeing evidence of “something” going on - from a personality disorder with a good prognosis to a clinical condition as serious as any personality disorder to a situation of wait-and-see to a pressing concern that merits some sort of intervention.

This may not be a perfect system, and already a predictably rotten tomato review has come in from Allen Frances, head of the criminally horrendous DSM-IV, who characterized Take Two as “an impossible mess to the rest of us.”

Dr Frances may well be right, but for all the wrong reasons. Dr Frances has indicated in all his DSM-5 writings to date that he sees himself as merely as the keeper of his precious DSM-IV, which is a very different proposition than lending his professional wisdom to improving the lives of those dealing with serious personality issues. In the final analysis, any attempt to pin down something as infinitely complex as personality is doomed to be flawed. Success, then, is modest, to be measured in terms of less flawed than the effort before.

Could the DSM-5 have done a better job? Of course it could have. Is the version it turned in way better than the sorry DSM-IV mess that Dr Frances is so in denial about? Don’t make me answer that.

Previous pieces:

Personality Disorders - The DSM-5 Has Another Go
Take Two on Personality Disorders

Personality Disorders - The DSM-5 Has Another Go

On June 21, the DSM-5 workgroup responsible for bringing order to personality disorders substantially changed what it originally posted on the DSM-5 site back in Feb, 2010. Most of us could see this coming. As the workgroup explained in its most recent update: “All parts of the model have been simplified and streamlined in response to comments received and to critiques in the published literature.”

A little background ...

Unlike the rest of the DSM-5, the 2010 roll-out of Personality Disorders was no mere light dusting of the status quo. This was a major home-improvement, a hugely ambitious effort that sought to integrate two different ways of looking at mental illness - categories (as in “which one?”) with dimensions (as in “how much?”) into something workable.

Personality, after all, cannot be easily categorized. Yet, we do need categorical bearings. Dimensionality is far more in sync with clinical reality, yet much more difficult to sort and freeze into comprehensible diagnostic nuggets. The workgroup’s answer to this dilemma was essentially a modular system. (The diagnostic literature refers to “hybrid,” but “modular” - think IKEA - is far more accurate.)

Thus, the DSM-5 retained diagnostic categories such as Borderline and Antisocial, but built them with interchangeable parts. Accordingly, individuals with either illness may have the following hostility “personality trait” in common:

Persistent or frequent angry feelings; anger or irritability in response to minor slights and insults ...

This particular trait is a subset of the antagonism “domain.” But quick comparison between the two categories reveals very different loadings. Those with antisocial have far more antagonism traits (such as callousness) than those with borderline. (The new version lists only one antagonism trait for borderline, down from two in the old version.) Meanwhile, borderline comes heavily loaded in negative affectivity traits (such as emotional lability) while we see nothing of the sort in antisocial.

Thus, a few shared traits, but clearly defined separation, which hopefully results in less diagnostic confusion.

That brings us to the dimensional element. In their initial roll-out, the DSM-5 essentially took apart and reassembled the exact same categorical bits and pieces in a different way to see “how much” of say borderline or antisocial (or avoidant or schizotypy or obsessive-compulsive) one has.

In other words, the real world is not black and white, much less divided into all-or-nothing choices. Imagine: Under the DSM-IV, a person who is one symptom shy for not only borderline, but also antisocial and narcissism is technically normal. How crazy is that? The DSM-5’s solution was to look at the entire gamut of personality as something analogous to blood pressure, from healthy to unhealthy, with an eye on the various loadings.

One may or may not require clinical attention, but a dimensional assessment can be an enormous help in knowing thyself.

Great concept, but the DSM-5’s first draft was monumentally unwieldy, calling for clinicians to rate 37 facets making up six domains on a four-point scale. Who has time for that?

It was back to the drawing board for the DSM-5 work group. Eighteen months later, the blood pressure principle remains the same, but now, in place of version number one, we have a new dimensional entity called Personality Disorder Trait Specified (PDTS).

Oh-oh! Are we in for yet another mouthful of diagnostic alphabet soup, this time looking like a misspelling of PTSD? Or is the DSM-5 about to make life easier for us?

More to come ...

Taking it Personally: The DSM-5 and the Narcissism Controversy

I'm back into mucking out my Augean Stables, otherwise known as updating my mcmanweb site. I have a number of great pieces (at least I think they're great) lined up for this blog, but, in the meantime, from the vaults of "Knowledge is Necessity" (Dec, 2010) ...

A few weeks ago, the NY Times featured a piece by psychologist Charles Zanor entitled, A Fate Narcissists Will Hate: Being Ignored. The long and short of it is that narcissistic personality disorder will be axed from the DSM when the next edition is published in 2013. Imagine how your average narcissist must feel.

The article mentions that retiring the diagnosis has drawn the wrath of clinicians, who view the various committees of the DSM-5 as dominated by academics out of touch with reality. John Gunderson of Harvard, one of the leading authorities on personality disorders, called the decision “unenlightened.”

Actually, overhauling the entire field of personality disorders is probably the only thing those charged with the DSM-5 did right, though with reservations. A little background:

In this post-Freud era of biological psychiatry, “Axis II” personality disorders have been accorded a lot less respect than “Axis I” disorders such as depression, bipolar, anxiety, or schizophrenia. A cynic would say that is because there are no meds for Axis II disorders and they would be one hundred percent right. The upside to this is there has been no big pharma to call the shots. It is no coincidence that the only major reforms to the next DSM occurred in the one realm where pharma is conspicuously absent.

The first obvious change is no Axis I/Axis II distinction. Personality disorders will get the same billing as mood disorders and anxiety disorders and all the rest. The next obvious change is a new “dimensional” component to complement the “categorical” classification of personality disorders, something that should have been done with mood disorders and arguably the whole rest of the DSM.

In its background papers and rationale, the APA and the DSM-5 group note that separating out personality into discrete illnesses has generated no end of end of clinical confusion. Is someone who abruptly breaks off a friendship, for instance, an “antisocial” with no remorse, a “borderline” who can’t cope, or a “narcissist” who cares only about him or herself?

Clinicians typically hedge their bets by choosing more than one, or by tacking on the NOS (not otherwise specified) qualifier.

The dimensional view acknowledges the complexity and subtlety of personality. Instead of asking “which one?” at the expense of ignoring whatever else may be going on, a clinician would be asking “how much” and “how severe?” In a sense, psychiatry is bringing back neurosis, but with some important refinements.

The personality disorders we are most familiar with are grouped into “Cluster B” in the current DSM. They include borderline personality disorder, antisocial personality disorder, narcissistic personality disorder, and histrionic personality disorder. The next edition of the DSM will give narcissistic personality disorder and histrionic personality disorder the boot, and prune out a number of other disorders from clusters A and C as well, leaving us with five “personality disorder types,” including antisocial/psychopathic, borderline, avoidant, obsessive-compulsive, and schizotypal.

According to the rationale provided in the draft DSM-5, three of these types have the “most extensive empirical evidence of validity and clinical utility.”

Here’s where the dimensional component would come in. The draft DSM-5 proposes testing for six “trait domains” that would include:

1. Negative Emotionality (such as depression or anxiety).
2. Introversion (such as social withdrawal and intimacy avoidance).
3. Antagonism (such as callousness, manipulativeness, narcissism, histrionism, hostility, aggression, oppositionality, and deceitfulness).
4. Disinhibition (such as impulsivity).
5. Compulsivity (such as risk aversion).
6. Schizotypy (involving odd behaviors and cognitions).

These trait domains are based on well-established personality tests such as the five-factor model and Cloninger’s psychobiological model, and would seek answers (note the plural) to such life mysteries as why an individual would abruptly break off a friendship.

Sounds good in theory, but are clinicians too set in their old ways? As the NY Times notes:

Clinicians like types. The idea of replacing the prototypic diagnosis of narcissistic personality disorder with a dimensional diagnosis like “personality disorder with narcissistic and manipulative traits” just doesn’t cut it.

Sounds a bit narcissistic to me, wait, I mean - uh - never mind.

Much more to come ...

Personality Disorder: Understanding Dimensionality

The story so far: The DSM-5, due out in 2013, will be the same old book in new covers. The exception is personality disorders, which gets a major overhaul. Not coincidentally, this is the one realm of mental illness where big pharma is conspicuously absent.

My last two blog pieces - Decisions, Decisions, and Bringing Order - looked at the changes on the “categorical” side of personality disorders. Five out of ten of the present disorders will get the boot, while the five left standing (antisocial/psychopathy, avoidant, borderline, obsessive-compulsive, and schizotypy) receive further clarification. The weakness with categories is inevitable overlap, but rather then pretending this doesn’t exist, the DSM-5 openly adopts interchangeable parts, with six “trait domains” (negative emotionality, introversion, antagonism, disinhibition, compulsivity, and schizotypy) further subdivided into “facets.”

Thus, the new borderline “type” (which replaces the term “disorder”) is loaded with six negative emotionality trait domains and two antagonism domains, while the new antisocial/psychopathy type is heavy on antagonism (six in all) and light on negative emotionality (zero, in fact). The two borderline antagonism facets (hostility and aggression) reappear word-for-word in antipsychocial/psychotic.

Kind of like IKEA. The parts may be the same, but each piece of furniture - one hopes - is very different. So what would happen if we were to dispose of the concept of furniture altogether? Funny you asked.

Enter the “dimensional” model. The same bits and pieces are there - namely the general trait domains of negative emotionality, introversion, antagonism, disinhibition, compulsivity, and schizotypy, plus the more specific facets - but instead of looking for labels, we are looking for shadings. Instead of asking “Which one?” (as in borderline or antisocial) we are asking “How many?” and “How much?”

The DSM-5 trait domains are derivative of, and roughly correspond to, the traits in the five-factor model (FFM), already in wide use in clinical practice. The FFM tests for “openness to experience”, “conscientiousness”, “extraversion”, “agreeableness”, and “neuroticism” (OCEAN).

A quick scan, however, reveals that the FFM and DSM-5 investigate essentially the same phenomenon from entirely different viewpoints (“extraversion” vs “introversion”, “agreeableness” vs “antagonism”, “conscientiousness” vs “disinhibition” and “compulsivity”). Whereas the FFM looks at our potential, the DSM looks at what is holding us back.

A good illustration of this is the FFM’s “openness to experience” trait, which has no apparent correspondence in the DSM-5. The opposite to openness to experience is stuck in the mud, which you can hardly associate with any kind of mental disorder. In this context, stuck in the mud is simply less desirable than being creative and intellectually adventurous.

Now let’s examine the one DSM-5 trait with no seeming FFM counterpart. “Schizotypy” broadly describes strange thinking and behavior. Its opposite? May I suggest stuck in the mud? In other words, in a DSM setting, stuck in the mud comes across as a desirable trait. Certainly, no one calls 911 to complain about their boring wife or husband.

The DSM-5 calls for clinicians to rate all 37 facets which make up the six domains on a four-point scale, which invites the instant criticism of clinical unwieldiness. According to Allen Frances, who chaired the DSM-IV, blogging on Psychology Today:

Unfortunately, the reach of DSM-5 far, far exceeds its grasp. Only by going to the website yourself and reviewing the DSM-5 dimensional suggestions can you get a feel for just how remarkably ad hoc, idiosyncratic, and cumbersome they are. I have discussed the suggestions for dimensional personality disorder ratings with a number of experts (and this is also my area) and none of us could decipher the proposal, much less conceive of its ever being workable. One described it as an example of "too many research cooks spoiling a clinical broth".

Dr Frances’ criticism may well be valid, but it also can be interpreted as an egregious case of DSM-worthy “lazy clinician syndrome.” We see this every day in doctors who profess to be far too busy to monitor their patients for weight and blood sugar levels and other red flags when prescribing meds with notoriously high metabolic risks, not to mention other high crimes and misdemeanors.

Personality is fiendishly time-consuming and complex. Clinicians want it quick and simple. But what about our interests?

Much more to come ...

Previous Pieces

Taking It Personally: The DSM-5 and the Narcissism Controversy

Let's Play Spot the Personality Disorder 

Why is Spotting the Personality Disorder So Damned Hard? 

Personality Disorders: Decisions, Decisions ...

Bringing Order to Personality Disorder

Taking it Personally: The DSM-5 and the Narcissism Controversy

A few weeks ago, the NY Times featured a piece by psychologist Charles Zanor entitled, A Fate Narcissists Will Hate: Being Ignored. The long and short of it is that narcissistic personality disorder will be axed from the DSM when the next edition is published in 2013. Imagine how your average narcissist must feel.

The article mentions that retiring the diagnosis has drawn the wrath of clinicians, who view the various committees of the DSM-5 as dominated by academics out of touch with reality. John Gunderson of Harvard, one of the leading authorities on personality disorders, called the decision “unenlightened.”

Actually, overhauling the entire field of personality disorders is probably the only thing those charged with the DSM-5 did right, though with reservations. A little background:

In this post-Freud era of biological psychiatry, “Axis II” personality disorders have been accorded a lot less respect than “Axis I” disorders such as depression, bipolar, anxiety, or schizophrenia. A cynic would say that is because there are no meds for Axis II disorders and they would be one hundred percent right. The upside to this is there has been no big pharma to call the shots. It is no coincidence that the only major reforms to the next DSM occurred in the one realm where pharma is conspicuously absent.

The first obvious change is no Axis I/Axis II distinction. Personality disorders will get the same billing as mood disorders and anxiety disorders and all the rest. The next obvious change is a new “dimensional” component to complement the “categorical” classification of personality disorders, something that should have been done with mood disorders and arguably the whole rest of the DSM.

In its background papers and rationale, the APA and the DSM-5 group note that separating out personality into discrete illnesses has generated no end of end of clinical confusion. Is someone who abruptly breaks off a friendship, for instance, an “antisocial” with no remorse, a “borderline” who can’t cope, or a “narcissist” who cares only about him or herself?

Clinicians typically hedge their bets by choosing more than one, or by tacking on the NOS (not otherwise specified) qualifier.

The dimensional view acknowledges the complexity and subtlety of personality. Instead of asking “which one?” at the expense of ignoring whatever else may be going on, a clinician would be asking “how much” and “how severe?” In a sense, psychiatry is bringing back neurosis, but with some important refinements.

The personality disorders we are most familiar with are grouped into “Cluster B” in the current DSM. They include borderline personality disorder, antisocial personality disorder, narcissistic personality disorder, and histrionic personality disorder. The next edition of the DSM will give narcissistic personality disorder and histrionic personality disorder the boot, and prune out a number of other disorders from clusters A and C as well, leaving us with five “personality disorder types,” including antisocial/psychopathic, borderline, avoidant, obsessive-compulsive, and schizotypal.

According to the rationale provided in the draft DSM-5, three of these types have the “most extensive empirical evidence of validity and clinical utility.”

Here’s where the dimensional component would come in. The draft DSM-5 proposes testing for six “trait domains” that would include:

1. Negative Emotionality (such as depression or anxiety).
2. Introversion (such as social withdrawal and intimacy avoidance).
3. Antagonism (such as callousness, manipulativeness, narcissism, histrionism, hostility, aggression, oppositionality, and deceitfulness).
4. Disinhibition (such as impulsivity).
5. Compulsivity (such as risk aversion).
6. Schizotypy (involving odd behaviors and cognitions).

These trait domains are based on well-established personality tests such as the five-factor model and Cloninger’s psychobiological model, and would seek answers (note the plural) to such life mysteries as why an individual would abruptly break off a friendship.

Sounds good in theory, but are clinicians too set in their old ways? As the NY Times notes:

Clinicians like types. The idea of replacing the prototypic diagnosis of narcissistic personality disorder with a dimensional diagnosis like “personality disorder with narcissistic and manipulative traits” just doesn’t cut it.

Sounds a bit narcissistic to me, wait, I mean - uh - never mind.

Much more to come ...